New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
I210 merge #839
I210 merge #839
Conversation
…into subnetwork_dev
…into subnetwork_dev
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, some minor comments
@@ -147,7 +147,6 @@ def setup_exps_rllib(flow_params, | |||
config["lambda"] = 0.97 | |||
config["kl_target"] = 0.02 | |||
config["num_sgd_iter"] = 10 | |||
config['clip_actions'] = False # FIXME(ev) temporary ray bug |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
is this fixed when we upgrade to 8.0.0?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
yes. 0.7.3 fixes it too
"Currently, multiagent experiments are only supported through "\ | ||
"RLlib. Try running this experiment using RLlib: 'python train.py EXP_CONFIG'" | ||
else: | ||
assert False, "Unable to find experiment config!" | ||
if flags.rl_trainer == "RLlib": | ||
if flags.rl_trainer.lower() == "rllib": |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
why rename it? just for aesthetics?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
They can just type whatever they want and capitalization errors wont throw an error
flow/core/experiment.py
Outdated
|
||
import logging | ||
from flow.core.util import emission_to_csv | ||
from flow.utils.registry import make_create_env | ||
import datetime | ||
import numpy as np | ||
import logging | ||
import time | ||
import os | ||
|
||
from flow.core.util import emission_to_csv | ||
from flow.utils.registry import make_create_env | ||
import matplotlib.pyplot as plt | ||
import numpy as np |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
are these ordered correctly?
flow/core/experiment.py
Outdated
custom_callables : [str, lambda] | ||
List of strings and lambda functions corresponding to some information we want | ||
to extract from the environment. The lambda will be called at each step to extract | ||
information from the env and it will be stored in a dict keyed by the str. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
maybe add something that they should return a numerical quantity. i think this script breaks if the callable returns None
STEPS = 10 | ||
rdelta = 255 / STEPS | ||
# smoothly go from red to green as the speed increases | ||
color_bins = [[int(255 - rdelta * i), int(rdelta * i), 0] for i in range(STEPS + 1)] |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
this is cool, but SUMO does this by default btw :^)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
How do I turn this on in SUMO?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can you replace this with the correct command?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
non-trivial to implement, have to add a gui-settings file and manually set that as the default setting for vehicle coloring
i.e. https://sumo.dlr.de/docs/SUMO-GUI.html#configuration_files
tests/fast_tests/test_examples.py
Outdated
@@ -247,6 +253,17 @@ def test_multi_highway(self): | |||
} | |||
self.run_exp(multiagent_highway, **kwargs) | |||
|
|||
# TODO(@evinitsky) enable |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Shouldn't leave this in master if it doesn't work
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Think it's alright, we have other tests like this that we need to turn on at some point. Not perfect code but this isn't a stable project till a version release anyways
still need to write tests but then we can probably merge |
What do you want to test before we merge this? |
08fed83
to
b5bfb65
Compare
54cd1b4
to
e6653e1
Compare
e6653e1
to
9367d96
Compare
@AboudyKreidieh I have tests for everything but param_sweep.py. Are you down to merge? |
ec64c8c
to
72ba4f4
Compare
72ba4f4
to
154441c
Compare
Merge the I210 network into master so everyone has access to it.